REST plugin invocation on multi-node cluster

s are discovered after changing elasticsearch.yml file . But for load balancing of custom plugin invocation is there any built-in plugin or elastic search feature available and it should also switch node when one node goes down Thanks, Chetana -- Please update your bookmarks! We have moved to

deploying elasticsearch on AWS

I have to deploy ES version 0.90.10 on AWS. For that I have installed cloud-aws plugin having the version 1.17.0-SNAPSHOT. But I want to use *local* gateway instead of S3 gateway. I have added/uncommented these 2 lines in elasticsearch.yml cloud.aws.discovery.type: ec2 gateway.type: local I

Queries on elasticsearch index template

I am using index template to store metadata of my application. There are many templates (including nested) created. 1. Is template a good idea for application metadata (not all are search related) storage? 2. Does ES store template details in each node or like index data is it possible to spl

restore api vs custom copy script

I have a secondary index storage where I have both snapshot (taken from ES server) and valid lucene index (content is same as snapshot). If ES server index is cleared/deleted though the mapping is retained, could someone tell me which is the better option to restore in terms of time taken: a)

efficient merging of index data stored in heterogeneous repositories

I have some index data stored in hdfs (ES compatible indexing) and rest in ES and also ES and hdfs clusters are different. Is there any other better alternative than below options 1. Bulk index request - but it is unnecessary as the data is already in indexed format 2. Move/copy periodically h

Re: is it possible to update snapshot with new lucene segment and restore

Could please suggest a best option for merging index data stored in HDFS with the index data stored in ES node On Saturday, May 31, 2014 1:45:13 AM UTC+5:30, Binh Ly wrote: > No I don't believe so. The snapshot data is not really a "valid" Lucene > index, per se. It does contain segment files,

is it possible to update snapshot with new lucene segment and restore

I have a requirment where for some data setting the '_source' in indexrequest is strightforward but for some huge amount of data I need to run long running activity to generate data which needs to be indexed. So frequently I plan to take a snapshot from ES to Hadoop and want to add new lucene

Re: TransportClient of ES 1.1.1 on Ubantu 12.x throws 'No node available' exception

witter : @dadoonet / @elasticsearchfr / @scrutmydocs > > > Le 28 mai 2014 à 10:34, Chetana > a > écrit : > > ES log shows this warning message, 'transport.netty- Message not fully > read (request) for [0] and action [], resetting' > On Wednesday, May 28, 2014 12:07:39

questions on custom routing

I want to create an index repo with multiple (3) shards and use explict/custom _routing while indexing. I have a couple of questions, can somebody please answer 1. If I explictly set _routing, does ES take care of load balancing of index data, indexrequest, search request

Re: TransportClient of ES 1.1.1 on Ubantu 12.x throws 'No node available' exception

ES log shows this warning message, 'transport.netty- Message not fully read (request) for [0] and action [], resetting' On Wednesday, May 28, 2014 12:07:39 PM UTC+5:30, Chetana wrote: > I have ugraded ES from 0.90.2 to ES 1.1.1 recently. I am able to get the > Tranpsort client

TransportClient of ES 1.1.1 on Ubantu 12.x throws 'No node available' exception

I have ugraded ES from 0.90.2 to ES 1.1.1 recently. I am able to get the Tranpsort client of ES 1.1.1 and create mapping on Windows 2010. But I get a 'No node available' exception on Ubantu 12.x Here is the code I use for connecting to ES ImmutableSettings.Builder settingsBuilder = Immutable

issue with client.admin().cluster().prepareGetSnapshots(...)

The call to prepareGetSnapshots(...) for getting a snapshot which is not existing, throws SnapshotMissingException. But I expect instead it should return response with a list of zero snapshots(getSnapshots()) or atleast isExist=false Is there any other way one can check the existence of a snap

how to take periodic snapshots

I am using ES 1.1.1 and hdfs as repository type (Hadoop Hortonworks 2.2.2). I need to take snapshot for every half hour. At present, I delete the old snapshot to take the new one to avoid 'snapshot already exists' exception. Is there a way to update a snapshot ?. If available, please provide

unable to take snapshot and restore using repository-hdfs

I am using ES 1.1.1 and Hadoop hortonworks 2.2.2 and have installed the plugin '*repository-hdfs' (https://github.com/elasticsearch/elasticsearch-hadoop/tree/master/repository-hdfs )* in ES. I have created one

Re: unable to index using elastic-hadoop plugin

r logging, make sure you properly configure log4j for your Hadoop > environment - it depends on what version you are > using and what libraries. > If the job is complete, you can always tests the results by querying ES > for the data that was just indexed. > > On 4/28/14 9:2

Re: unable to index using elastic-hadoop plugin

ransmitted, > the reply, etc > > On 4/24/14 12:21 PM, Chetana wrote: > > Yes, I am able to search but not able to index. I am using 1.3.0.M2. The > json file size is just 28bytes. > > I corrected port/host and some log setting. With all these chnages I am > to

Re: unable to index using elastic-hadoop plugin

easier to > use and still the one the vendors recommend by default > - since you are using Hadoop 2 from Hortonworks, you might want to upgrade > to their latest HDP (2.1) release. > > Hope this helps, > > [1] http://www.elasticsearch.org/overview/marvel/ > [2] > http

unable to index using elastic-hadoop plugin

I am using elasticsearch-hadoop plugin ( https://github.com/elasticsearch/elasticsearch-hadoop) and trying to index some documents. Iam using elasticsearch version 0.90.2 and Hadoop Hortonworks 2.2.0. The search functionality works fine, but while indexing application hangs https://github.com/

need help for Hadoop and ES integration

I have downloaded elasticsearch-hadoop-1.2.0.jar from github and trying to search. The code to search looks like * Configuration conf = new Configuration(); conf.set("es.nodes", "localhost"); conf.set("es.port", "9200"); conf.set("es.resource", "test_index/test_mapping"); conf.se

search cascading in ES

We are developing an application which requires cascaded (flow based) search where the search result of one will become the input criteria for the next search. Is there a way to do this in ES ? If not, can you suggest some third party library which can provide cascading functionality over ES s

choosing shard vs alias in elasticsearch

Sorry it is alias not alice On Friday, March 21, 2014 10:11:53 AM UTC+5:30, Chetana wrote: > I am planning to use elasticsearch (ES) for storing event logs. Per day, > the application should store nearly 3000+ events and size will be around > 30-50K. > I need to take som

choosing shard vs alice in elasticsearch

I am planning to use elasticsearch (ES) for storing event logs. Per day, the application should store nearly 3000+ events and size will be around 30-50K. I need to take some statistics monthly, half yearly, yearly also year old data can be ignored sometimes but data should be retained for

Re: facets.total and hits.total dont match

I have indexed some records by making test_field to be 'analyzed'. If the analyzed field causing this issue, is there any other facet type/work around which can solve the problem? On Tuesday, January 21, 2014 12:15:45 PM UTC+5:30, Chetana wrote: > I have an application wher

facets.total and hits.total dont match

I have an application where I need both search results and facet information. Everytime a query is framed based on some filter condition and query words and it is passed to both facet and search request as given below. The field (test_field) on which the facet to be applied is present in all do

Re: Case (lower and upper) issue with TermsFacetBuilder result

I have observed this behavior on Ubantu 11.10 version. On Wednesday, January 15, 2014 10:29:41 AM UTC+5:30, Chetana wrote: > I am using the 'term facet' functionality of ElasticSearch 0.90.2 on a > paticular field (datatype). The field (datatype) stores one of the enum > v

Case (lower and upper) issue with TermsFacetBuilder result

I am using the 'term facet' functionality of ElasticSearch 0.90.2 on a paticular field (datatype). The field (datatype) stores one of the enum values in camel case (example : sampleType). I am using a standard analyzer for index and search which internally uses lowercase filter. The strange th

unexpected behavior of pagination using offset and size

My application has the pagination requirement for search. I am using the offset and size option to achieve the pagination. Making quick clicks on pagination sometimes does not give results at all. Does the asynchronous search call bringing any side effects like this? Thanks, -- You recei

issues with timestamp sorting

I have created a timestamp field (not be confused with _timestamp) and want to sort timestamp field in descending order. But the result contains some records out of order. Mapping and the sorting criteria look like : "timestamp":{"type" :"date","format":"dateOptionalTime","include_in_all":fals

facet_filter for boolean field does not work in TermsFacetBuilder

I have a boolean field 'flag' which I want to filter for facet field 'sub_type'. But the total count is zero though there are some documents satisfying the criteria. But if I apply facet-filter on any other field (non-boolean) I get the results. Is there any work around for this? Json of Face

Is Percolate with multiterms possible in ES

I have a page with different sections say abstract, 'see also' etc. I want to register queries with some word appearing in abstract field and some other word appearing in 'see also' field So my registerer query looks like : String source = {\"query\": {\"term\":{ \"abstract\" :\"laptop\", \"se